Roundup: 3 New Beginner DSLRs

[dwu1020]

"Then why are non-EF-S or non-Nikkor AF-S/AF-I lenses mentioned?"

This should read "why aren't non-EF-S..."

 

[Guest]

I may be misunderstanding something, but on the Color & Quality page, you say that a higher Signal-to-Noise ratio is bad. Maybe I've missed your definition of the term, but higher SNR is usually better. Also, the Olympus, with by far the smallest sensor, got the lowest SNR. I looked at the images (histogram ones) and there isn't a noticeable difference between the Nikon and Canon, but there is noticable noise in the Olympus. Great review otherwise, I'm glad you guys are talking about Cameras.

 

[Guest]

After reading the article again, I have to retract the "Great review" bit of my comment. The sheer wrong-ness of tying Depth of Field to number of Autofocus points means whoever wrote this review is unqualified to do so. Whether this article is meant for beginners or not, this is inexcusable. People trust the Tom's brand and I don't think you guys should keep doing camera reviews until you have a qualified reviewer in-house.

 

[zodiacfml]

haplo602 beat me to it, yeah, he's right about links and articles regarding each technical aspect and features. being a beginner, i don't know what depth of field, SNR, ISO, and f-stops are.
regarding taking HD videos, i think, it's a useful feature for non-pros like me.
i like to see a larger features comparison table.

in my opinion, Scores should be placed at the bottom page, or better, at the end of article, not in between sentences.

regarding image quality, it is difficult to trust and understand an author's opinion about it without a side by side comparison of the same cropped images of the said cameras. (no need for zoom link)

lastly, there are a lot of redundant or not quite important parts such as advices like use of strap, bring two batteries, and etc.

 

[haplo602]

[citation][nom]andrewho[/nom]I may be misunderstanding something, but on the Color & Quality page, you say that a higher Signal-to-Noise ratio is bad. Maybe I've missed your definition of the term, but higher SNR is usually better. Also, the Olympus, with by far the smallest sensor, got the lowest SNR. I looked at the images (histogram ones) and there isn't a noticeable difference between the Nikon and Canon, but there is noticable noise in the Olympus. Great review otherwise, I'm glad you guys are talking about Cameras.[/citation]

ah that's a good point. I did not actualy noticed that they take higher SNR as worse. I was just puzzled why the Olympus is the best while the actual images are the worst from ISO 400 up.

Actualy I'd like to know how they measure the SNR, might be interesting.

One point for the color/image part. Nikon underexposing images is actualy a +. You can always make the image brighter, but if you overexpose, there's nothing you can do to salvage the image (not with digital).

One more point, while all of them offer a RAW format, Nikon makes you pay for their raw processing software while Canon gives it for free (I have no clue about Olympus). Also Canon gives a free camere control utility that you can use to shoot/control the camera from you laptop/PC.

 

[kartu]

[citation][nom]johanncoet[/nom]I find it very disappointing that Rick Oldano did not even mention once in this review that the Canon T1i has a HD Video function. It can shoot Full HD video at 1080p (only at 20fps which is kind useless thought) and 720p at full 30fps with amazing quality. It is one of only a few DSLR cameras that can shoot video. And it equals or even beat other models like the Nikon D90. This is one of it main selling point and makes it a much better competitor. I would like to know why it was not mentioned as this was a review of the overall functionality of the three cameras and this is a pretty big function. Just thought I would let you guys know because I would think it important in deciding between the three.[/citation]
Who, besides Canon's marketing department, needs 1080p20?!?!?!?

First SLR to shoot HD (720p) was Nikon D90. D5000 also shoots video.

 

[Tomsguiderachel]

[citation][nom]samihaha[/nom]I guess I will have to apologize. I am not aware of the distinction between Tom's hardware and Tom's guide. I was directed to this article from the homepage of toms hardware and assumed the articles are written by the same people following the same standard. Misconception might be that I considered articles published under "Tom's" brand are substitutes for professional publications. Sounds like not all "Tom's" are created equal. Thanks for pointing that out.[/citation]
That's right--the sites are not equal. One covers computer parts only, the other covers consumer electronics. Hardware covers computer parts very in-depth, testable, while Guide covers consumer electronics for the beginner.

Thanks for understanding,
Rachel

 

[just0a0guy]

with the canon lens selection it is important to point out the EF-S line is meant for non full frame sensors and will not work on a film camera or top of the line full frame digital camera. the EF line can be used on full frame digital or film cameras with an EF mount and all crop sensors(which has been in use for over 9 years now, my film camera bought in 2000 uses the same lens mount making it very attractive to pay more for glass and use it on two bodies). the nikon compatibility was already address in another post but again it is important to let advanced novices know what to expect from any previous investment that may have and what to expect when they purchase new glass given that any one who feels like more then a novice may be intending to grow the skill set and might be looking into an investment that will last longer then the 18-24 months of a point and shoot.

features and design: the physical size was already mentioned in a previous post but it is important how it fits into your hand so letting the reader know about the size of the camera body is important. the other thing of great importance is the weight of the camera sort of nice to know if your lens will be heavier then the body or if you want a light weight body for backpacking when every pound counts.

color and quality: you may want to let the reader know what color depth is being used in this case 14 bit. also for a fair comparison make sure all cameras are using the same white balance selection i.e daylight or indoor, tungsten ect..

price: please always list body only price and kit (body+lens) price

Summary:
you mention that you have concerns about the canon go ahead and remind us what they are because aside from the brightness i don't recall you mentioning any other items about it negatively. Also please add a list of any software that ships with the camera. while most people already have some application they are using sometimes it makes a world of difference if they are using outdated software. cameras at least a dslr is a bit mroe of an investment then a point and shoot so the upgrade cycle might be years in between resulting in very old software with alot to be gained from the latest version being included with the camera.

 

[johanncoet]

Hi Rachel
Thankyou for clearing this up, I wasn't aware of the difference between tomshardware and tomsguide. I complained about the lack of mentioning the HD Video feature but I should mention that it is included in a different review available from tomshardware. So I guess people can just look at that review aswell.

Thanks for your response.

 

[rloldano]

Several readers have voiced questions and opinions ranging from "Can you clarify this" to comments bordering on "He doesn't know what he's talking about," regarding my review of three new single-lens-reflex cameras. Let me try to answer some of their questions. Bear in mind that Tom's Guide is not DPReview (www.dpreview.com) or Imaging Resource (www.imaging-resource.com). We do not generate 7,000 to 10,000 word reviews that can cover the smallest details and features of any one camera. As Rachel Rosmarin commented earlier this week, if you need that depth in a review, you will need to turn to other resources.

One reader wrote that you should "forget the macro modes on P&S cameras, they are all about the closest focusing distance." That's quite true. Unfortunately, the reader misses the point of these cameras. A large number of the consumers purchasing these cameras are buying them as their second or third digital camera. They have, in fact, been using a point & shoot camera with the macro mode. These currently reviewed SLR cameras can be seen as a follow-on purchase for these valued customers. You may have been familiar with 35 mm film SLR cameras, and then migrated to digital via the point & shoot camera. As a vendor, Canon or Nikon or Olympus know very well that they will never secure large sales of their "prosumer" and professional digital SLRs as compared to their film SLRs, not when that equivalent costs north of $3,000 to $ 8,000.

So, why should these cameras even have a "Macro" feature? Think about it: You have just spent over $800 on one of these camera kits. The reason someone likely buys one of these cameras is for the availability of the additional interchangeable lenses. But, do you now want to force that buyer to purchase an additional lens just to take a macro shot, a lens that will set them back just under $400 for the Canon EF-S 60 mm macro lens, or around $450 for either the Nikon or the Olympus. Do you really want to be scaring off that new devotee?

Another point raised was that the focusing points of each camera's auto-focus system had nothing to do with depth-of-field. Once again, the reader is correct, the focusing points have absolutely nothing to do with depth-of-field. What the reader failed to address is that depth-of-field has everything to do with where you focus and the ƒ-stop used for that image. For example, you frame an image and focus on a subject six feet in front of you, using an ƒ-stop of ƒ 2.8. That subject will be in sharp focus, but everything beyond ten feet behind the subject will likely be out of focus or have a soft appearance.

But, let us say that it is really important for you to have everything from six feet to thirty feet in sharp focus. If you can use an ƒ-stop of ƒ 8 or ƒ 11 and focus on a point roughly in the middle of those two distances, you have a considerably better chance to get the picture you want. What if that center focus point is going to target someone in the middle of the picture, standing at six feet away? Look in your viewfinder, find an object that may rest roughly half way between the two distances. See if one of the other focusing points rests on that object. Select that point to force the camera to focus on that object. If that object is, in fact, about half way between the two distances, you stand a good chance that your image will have sufficient depth-of-field to cover the entire subject matter.

What is the fuss with multiplier effect some readers wanted to know. One reader wrote: " And again wrong. Yes, the lenses have a reduced image circle to match the smaller imager…" Which is exactly the point I was trying to make, nothing more. This reader went on to say that you would still need to use the multiplier effect to get the "true" focal length. My question must be, what is a "true" focal length? The "true" focal length, as I understand it, for these three cameras is the stated focal length of the supplied lenses. I do not need to multiply them to find out the focal length for a similar lens on a camera body first designed for 35 mm film.

One reader voiced a question regarding the measurement I used for determining an images' signal-to-noise ratio. It was this reader's understanding that the higher a signal-to-noise (SNR) ratio, the better. I have no background in electrical engineering, therefore as I understand it, I cannot give you the text book definition of what is, or is not, a perfect SNR.

What I can offer to you is the explanation I received from two Stanford University PhD professors, color scientists, who showed me how to asses an image from any camera or scanner, to determine how much electronic noise the image contained, thus determining whether that device would generate "clean" images.

Whether camera or scanner, create an image using some recognized form of grayscale. In this case, for these three cameras, I used a Macbeth ColorChecker, a photographic industry standard for several decades. On the bottom row of this card is a grayscale, ranging from white to black in six steps, or wedges. I then took a picture of this test target with each camera set to every ISO setting, low to high, in this case from ISO 100 for the Canon and Olympus and ISO 200 for the Nikon to ISO 3200. I do not test the "Lower" or "Higher" settings that some cameras now feature.

With the image open in Adobe's Photoshop application, I select the Marquee tool (the marching ants?). I draw several small marquees in the neutral gray, in this case, by my convention, the third square from the right side. Next I open the Histogram palette. As you can see in the Color & Quality section of the article, I have an illustration of just that, one for each camera, using the image set to ISO 100 or 200. As it was explained to me, the "STD. DEV." value, or the Standard Deviation, will define the signal-to-noise ratio. The higher the value, the "noisier" the image. I then average the several readings.

How do I know I am looking at a "noisy" image? In those images from the Macbeth, the gray wedge will appear to have a splotchy or mottled appearance. In that wedge, the splotchiness will take on the appearance of some pixels having a darker shade of gray while lying next to other pixels with a lighter shade of gray.

In a picture containing a significant amount of blue sky, you will see similar treatment of the blue, with lighter and darker pixels. In a portrait, skin tone will appear mottled, with a very "noisy" image looking like someone has a bad skin condition.

 

[Crabstick]

[citation][nom]rloldano[/nom]Bear in mind that Tom's Guide is not DPReview (www.dpreview.com) or Imaging Resource (www.imaging-resource.com).[/citation]

True, but it is no excuse to mislead people with inaccurate terminology. At least get the facts straight.

[citation][nom]rloldano[/nom]... But, do you now want to force that buyer to purchase an additional lens just to take a macro shot, a lens that will set them back just under $400 for the Canon EF-S 60 mm macro lens, or around $450 for either the Nikon or the Olympus. Do you really want to be scaring off that new devotee?[/citation]

You must do that, because otherwise the reader will get false impression that the macro mode would really be a true macro.

Regarding the DOF and focus point dilemma, it's is nice that you have corrected your self, but it must be corrected in the article also. And note, that the amount of focus points still has nothing to do with DOF. You achieve the same result with only 1 focus point. Just prefocus where you want and then reframe the shot. Basic camera usage.

[citation][nom]rloldano[/nom]What is the fuss with multiplier effect some readers wanted to know.[/citation]

You still refuse to understand the whole issue. Maybe it is not about the "true" focal length, but more of a comparable focal length. For every camera-lens-combination it is good to know the "35 mm equivalent focal length" of the used lens. This just is how people compare the focal lengths. The lenses carry the "real" focal length on them. Then you have to multiply that with the crop-factor x to get the 35mm equiv length.

 

[dwu1020]

[citation][nom]rloldano[/nom]See if one of the other focusing points rests on that object. Select that point to force the camera to focus on that object. If that object is, in fact, about half way between the two distances, you stand a good chance that your image will have sufficient depth-of-field to cover the entire subject matter.[/citation]

It is often much quicker (and more reliable) to manually pre-focus and re-frame than to use the camera's interface to manually select the desired focal point. If you want to keep a range of subjects in varying distances in focus by picking a focal point in a vague space between them, all the autofocus points in the world won't decipher your intent and pick a spot in a vague space. More autofocus points simply increase the ability of the camera to select the correct focal subject regardless of where it is in the frame. Depth of field is simply not an appropriate subject to tie to the matter of autofocus.

[citation][nom]rloldano[/nom]What is the fuss with multiplier effect some readers wanted to know. One reader wrote: " And again wrong. Yes, the lenses have a reduced image circle to match the smaller imager…" Which is exactly the point I was trying to make, nothing more. This reader went on to say that you would still need to use the multiplier effect to get the "true" focal length. My question must be, what is a "true" focal length?[/citation]

The actual focal length is that which is marked on the lens. However, for anyone who has used film SLRs before, they are accustomed to the magnification ratio that certain focal lengths provide. In that respect, a 50mm lens on a film SLR will provide the same magnification ratio as a 33-35mm lens on a digital SLR (varies by brand). 18mm on a film SLR would be an extreme fisheye, whereas 18mm on a digital SLR is merely "wide-angle." That's why 50mm prime lenses are so common... they were designed for film SLRs. However, on a digital SLR, they are actually a mild telephoto, and not nearly as useful. Given the long history of film SLRs, 35mm-equivalent focal-length designations are very very useful.

[citation][nom]rloldano[/nom]One reader voiced a question regarding the measurement I used for determining an images' signal-to-noise ratio. It was this reader's understanding that the higher a signal-to-noise (SNR) ratio, the better. I have no background in electrical engineering, therefore as I understand it, I cannot give you the text book definition of what is, or is not, a perfect SNR... As it was explained to me, the "STD. DEV." value, or the Standard Deviation, will define the signal-to-noise ratio. The higher the value, the "noisier" the image.[/citation]

You have completely confused standard deviation and signal-to-noise ratio. They are correlated, yes, but inversely so. A low std dev is good, whereas a high snr ratio is also good. Think about it... you want "signal" to be much higher than "noise." However, you want the distribution of color in a supposedly flat color section of an image to be very tight, i.e. low std dev. So, the two terms are inversely correlated, and absolutely NOT interchangeable.

 

[Tomsguiderachel]

Dwu1020,
Now you've confused me: Rick Oldano says that a high standard deviation is bad, because it means a bad signal to noise ratio (too much noise). Yet you seem to be saying the same thing. Yes or no?

Is your complaint that in the article, Rick has assigned a SNR value to a camera, which you are saying is actually the STD DEV value? If so, I see why you are confused. I think a discussion of SNR and STD DEV is too advanced for this article, personally. I think next time Rick won't share the STD DEV values in the article at the point where he attempts to discuss SNR.

Rachel

 

[jwl3]

Under "Features and Design" the only feature you decided to rate was the size of the grip???? You gotta be kidding me.

 

[jwl3]

Under "Features and Design" the only feature you decided to rate was the size of the grip???? You gotta be kidding me.

 

[Tomsguiderachel]

[citation][nom]JWL3[/nom]Under "Features and Design" the only feature you decided to rate was the size of the grip???? You gotta be kidding me.[/citation]
What made you think that? I think the writer took into account lots of features and design aspects, but the cameras only truly differentiated in the ergonomics of the grip--that's where the writer found a substantive difference, in his opinion.

 

[jwl3]

One of the absolute worst product reviews I've ever read. Grip size??? Talking about articulated screens without mentioning which camera had the best screen?

There was nothing substantive in this review except re: the color reproduction and even that was very lightweight. Dude, a Best Buy sales rep could have done better.

 

[dwu1020]

[citation][nom]Tomsguiderachel[/nom]Dwu1020,Now you've confused me: Rick Oldano says that a high standard deviation is bad, because it means a bad signal to noise ratio (too much noise). Yet you seem to be saying the same thing. Yes or no?Is your complaint that in the article, Rick has assigned a SNR value to a camera, which you are saying is actually the STD DEV value? If so, I see why you are confused. I think a discussion of SNR and STD DEV is too advanced for this article, personally. I think next time Rick won't share the STD DEV values in the article at the point where he attempts to discuss SNR.Rachel[/citation]

I agree with the author if by snr, he actually meant std dev. If the numbers he was presenting was std dev, he is correct, but using the wrong term. Either number is a good, if simplified, way of presenting a comparison of noise levels, but it is important specify the correct term. I was giving him the benefit of the doubt regarding the terms, which is why I did not comment on it earlier until other people brought it up.

 

[haplo602]

[citation][nom]rloldano[/nom]... authors comment ...[/citation]

OK I give up

Come on, you have to know better than this. First you go on about how a macro mode is beneficial to the novice DSLR user and then in the next section your are discussing focusing and depth of field decisions that are way beyond novice DSLR users knowledge. You have to keep consistent at some point.

If you noticed my macro comment (and the follow up) then you failed to grasp the point I was making (ok maybe my English is just that bad). Macro is all about the final picture size on the imager. If subject to picture size is 1:1 then we talk macro. Everything else is NOT macro (technicaly 2:1 and such is also macro ). Each lens features a maximum magnification at closest focus distance. Typicaly it's 1:2 to 1:7 for the normal lenses. A 'macro' camera mode may attempt to focus the lens to the closes focusing distance to offer the best magnification the lens is capable, but without a true macro lens, this will NEVER be macro. Also given how phase-change focusing works, autofocus in the usual scenarion for closest focusing distances will never work reliably on these cheap bodies.

The marketing strategy behind a macro scene mode on a DSLR is one thing, how the real worls works is a different one. And this is something I'd expect to see explained in a review. If you don't have the space (like dpreview), then don't mention features you cannot explain correctly, let the reader figure out by himself. Concentrate on points you can explain (and have the knowledge and means to do so).

Oh and about the focal length multplier. Let's picture a John Doe comming to Tom's guide researching his next buying decision. He wants a light and small DSLR and likes to take wide angle nature shots. He can have:

Nikon D5000 + 17-55 kit
Canon 500D + 17-55 kit
Olympus 620 + 14-40 kit

Do you think he can make a decision based on your review ? the 14-40 looks like the most wide angle lens. However all 3 have equal field of view (35mm focal length) with Olympus having a different aspect ratio.

So the reader wasted some time on your article without getting any usefull information to help him.

 

[haplo602]

[citation][nom]Tomsguiderachel[/nom]Dwu1020,Now you've confused me: Rick Oldano says that a high standard deviation is bad, because it means a bad signal to noise ratio (too much noise). Yet you seem to be saying the same thing. Yes or no?Is your complaint that in the article, Rick has assigned a SNR value to a camera, which you are saying is actually the STD DEV value? If so, I see why you are confused. I think a discussion of SNR and STD DEV is too advanced for this article, personally. I think next time Rick won't share the STD DEV values in the article at the point where he attempts to discuss SNR.Rachel[/citation]

I'll make it even worse. He is measuring noise performance on a plain colored surface where even heavy noise reduction will work great. However in actual real life images, this NR procedure will destroy detail, saturation and contrast. F.e. photographing a green patch on a color checker and a field of grass will yield vastly different results. The green color checker patch will produce great results, however the field of grass will end as a bunch of green blotches created by the heavy NR process.